LLM Guard

Validates LLM text output against a JSON Schema (ajv). Invalid output routes to a Repair output with a ready-to-inject repair prompt — wire it back to your LLM node to build a visible retry loop. Never throws.

Node Analysis Pending

This node's source code has not been analyzed yet. Our service processes nodes in a first-come-first-served manner, and it takes time to analyze the large number of community nodes.

While waiting for the analysis, you can:

Visit node's website
Review node's source code on NPM
Check the node's package documentation

LLM GuardInstall

Node Analysis Pending

Discussion

LLM Guard